To The Moon

The analyst move that caught my attention most was Broadcom’s upgrade. It highlights an important shift in the AI investment story: the opportunity is no longer limited to GPU makers. Broadcom is positioned across several critical parts of the AI infrastructure chain, including custom accelerators, networking and connectivity. As hyperscalers continue pouring billions into data centers, demand for customized chips and high-speed networking could become increasingly important. What I find particularly interesting is the economics. Cloud giants want better performance and lower costs, while custom silicon gives them more control over their AI workloads. That creates a potentially powerful second engine of growth for Broadcom. The bigger takeaway? AI infrastructure is becoming an ecosystem,

Disclaimer: Nothing I say or post should be considered financial advice. Please do your own due diligence before making any investment decisions.‌ So OpenAI announced a breakthrough with their latest model, Astra. It has capability to control your whole computer system for you, automating all the programs you have. This will improve efficiency and any nagging issues with processing of apps. However, this means you are subjecting the information on your computer to a third party AI model, who knows what it will do with it. So the issue with privacy and PDPA actually will deter me from using this. From the beginning of the year, two AI companies are thrusted to the forefront, OpenAI and Anthropic. They are seen as the frontrunners in this rat race, and most optimistic to IPO at end of 2026 o

The memory trade enters the week of 8 September with something it did not have a week ago: confirmation. SNDK enters around $1,740. MU around $1,017. Both have shown extraordinary relative strength. But from here, I think their paths diverge. SNDK has a new mechanical catalyst. MU has the cleaner fundamental catalyst. And this week, both have to survive a major macro test. 🔴 SNDK: The Countdown to 21 September Begins SanDisk will enter the S&P 100 on 21 September. That matters because index-tracking funds will need to reposition around the rebalance. But I would not confuse that with unlimited upside. The index catalyst is real but temporary. SNDK still needs NAND pricing, AI storage demand and its long-term customer commitments to justify the valuation once those flows are finished. M

Michael Burry Calls Lululemon His "Trickster"—Is LULU a Fat Pitch or a Value Trap? $Lululemon Athletica(LULU)$   fell 17.4% to $100.61 last Friday, its lowest level in roughly eight years, bringing its year-to-date decline to about 52%. A stronger U.S. jobs report pushed rate-hike expectations higher and weighed on consumer discretionary stocks, but Lululemon's much steeper drop was mostly company-specific: North American sales weakened further, full-year guidance was cut again, and international growth also started to slow. Even contrarian investor Michael Burry has become more cautious on the stock. So the real question is no longer how far LULU has fallen. It is this: At around $100, is Lululemon finally

My top pick is $Oracle(ORCL)$ . I’m watching it closely because AI and cloud infrastructure demand remain strong growth drivers, while the market is waiting to see whether its huge backlog can translate into real revenue and EPS growth. I’ll be watching for an EPS beat and continued cloud growth. If Oracle can show that AI demand is converting into actual bookings and revenue, I think the stock could regain momentum after its recent weakness. A strong outlook or guidance upgrade could be an even bigger catalyst than the headline EPS beat. For me, ORCL offers a better growth opportunity than chasing dividend yield. I wouldn’t chase a pre-earnings rally, but I’d c

🌟🌟🌟Why $Apple(AAPL)$ moat will stabilise:  While $NVIDIA(NVDA)$ builds the factories of the future, $Apple(AAPL)$ still owns the digital real estate where humans live.  NVIDIA creates the intelligence but Apple delivers it to over 2 billion active devices.  For the average person, AI is useless until it is woven into the fabric of his life like his morning routine, messaging apps & privacy boundaries. Apple's moat is built on its unique ecosystem.   By introducing localised deeply private AI, Apple makes it even harder for users to leave.  Apple is not selling raw compute.  Apple is selling convenience for its users. My

🌟🌟🌟I believe that the memory giants $SK hynix(SKHY)$ $Micron Technology(MU)$ & Samsung Electronics wil continue to stay relevant for the long haul as they are orchestrating a memory supercycle that is breaking the historical boom and bust cycle. Why? The HBM Inelasticity Wall: AI starves without processing speed.  The insatiable race to build larger LLMs has forced a massive hardware shift towards HBM.  HBMs takes 4 times longer to manufacture than a standard RAM.  Giants like Hynix & Micron are completely sold out through 2027. The Big Contracts: Hyperscalers are locked into Multi year prepayments just to guarantee allocations.  This means the memory giants have a irreplaceab

$Singtel(Z74.SI)$    Closed well at 4.51 on 26 August, likely to continue to rise higher towards 4.70. Yesterday, closed well at 4.49, this is rather bullish! Likely to rise up to test 4.68-4.70. Beyond 4.70, she may rise up further towards 4.90-4.92.Pls dyodd. SingTel - She is gaining strength likely to rise up to test 4.46 and above . Beyond 4.46, she may rise up further towards 4.70. Pls dyodd. 30 July 2026: Nice Green candlesticks spotted on the chart. Long time didnt see the price is up 20 cents to close at 4.44, Looks rather bullish. She may rise up to test 4.70 than 5.00. 1st quarter results is out. – Singtel on Aug 13 reported a 71.6 per cent fall in first-quarter net profit to $818 million from $2.88 billion the previous year.

For me, the biggest thing to watch is whether $Apple(AAPL)$ can turn the foldable iPhone into a genuine new upgrade cycle. It needs to be compelling enough to support higher ASPs and attract customers back to upgrading, rather than becoming just another niche premium product. I’m also watching AI closely. Apple doesn’t need to lead in foundation models, but AI must become useful enough to drive upgrades. Rising memory and chip costs also make pricing and margins increasingly important, especially as AI increases memory requirements. Overall, I see this event as John Ternus’ first major test. If Apple delivers on foldables, AI and margins, the market could start pricing in a new growth cycle—not just another iPhone refresh, which could be a meanin

Iggy's Journal: Haze, a Quiet Pullback, and an Oil Story Worth Watching 8 September 2026, AM Market Data Before markets, a quick note on the air. Singapore's back in transboundary haze territory, smoke from fires in Sumatra and West Kalimantan drifting over. PSI briefly crossed into Unhealthy in the central region on 4 and 5 September, eased back to Moderate, 63 to 85, by 7 September. NEA's advisories are daily right now, and with dry weather expected to hold over both Singapore and the fire areas, it could swing again if the wind cooperates the wrong way. On to markets. STI gave back a little of Friday's record close, down 0.17% to 5,792.28, mild profit taking in the banks, DBS off slightly to S$78.47, OCBC down 1.21% to S$31.88, while Keppel Corp gained 2.17% to S$11.79. Nothing dr

OpenAI and Anthropic Just Reignited the AI Hardware Trade The AI trade is rotating back toward hardware. OpenAI's GPT-6 Astra launch and Anthropic's latest research breakthroughs have helped revive a simple investment thesis: frontier AI progress still requires enormous physical infrastructure. Better AI still needs more compute GPT-6 Astra delivered another major jump in reasoning and scientific capabilities. More important for hardware investors, $NVIDIA(NVDA)$   CEO Jensen Huang said Astra was trained on more than 100,000 Nvidia Grace Blackwell NVLink72 GPUs and that another 400,000 GPUs are "coming online next." Anthropic is sending a similar capability signal, with Claude recently advan

Tigers… the real trade isn’t just $Apple(AAPL)$  stock Tomorrow’s Apple event (Sept 9) isn’t another incremental upgrade cycle. It’s the moment Apple finally goes after the ultra-premium customer with a device that could start north of SGD $3,000 (≈ US$2,370+). Rumours are converging hard: • iPhone 18 Pro / Pro Max — refined, powerful, but still “just” phones. • The real star: Apple’s first foldable (widely expected as iPhone Ultra or Fold). Book-style. ~7.8-inch inner display. Outer cover screen. Crease-minimised OLED. A20-class silicon. Touch ID power button. Pricing whispers start around US$2,000–2,500 (≈ SGD $2,530–3,160) and climb higher for big storage configs — easily clearing SGD $3,000+ (≈ US$2,370+) in Singapore, with top-end units